![](https://lemmy.blahaj.zone/pictrs/image/d360715b-b71d-4703-80e0-c039a384a548.png)
![](https://lemmy.blahaj.zone/pictrs/image/c32272f0-9123-4f62-9323-60aa46ed92c2.webp)
What about a bot that automatically removes post/comments and bans people from other instances from the community?
A bisexual nonbinary poster of memes and other things • They/Any
What about a bot that automatically removes post/comments and bans people from other instances from the community?
That’s a possibility. I would be concerned that the false positive rate is so high compared to the rate of actual CSAM that the FBI would just block anyone using this for reporting as spam.
What might be done is to track the detection rate of users. If anyone is significantly higher than the average they might be uploading CSAM. Only issue I see with this is the detector doesn’t have an equal false positive rate across all content. It could be that the user just uploads pictures of their kids playing at the park a lot.
A 1% false positive rate is probably going to be to high to reliability report every positive to the FBI. The rate of actual CSAM is likely to be much lower than this. If it’s 1 in 10,000 uploads, you will have 100 false positives and 1 true positive.
It seems that locally hosted images are down. That’ll be every uploaded image, thumbnails, profile image. Weird thing is that community icons and banners are still loading.
From my understanding of how federation and communities work is that the instance hosting a given community receives the postal, comments and votes for that community from other instances and then sends the combined data out to other instances that requests that data. Users from instances that aren’t federated could still interact if they both used a community on a third party instance they were both federated with. But I could be wrong about all that.
And your right. Brigading is probably too strong of a word outside of evidence of coordinating action.
Now that I’ve looked at lemmy.world, beehaw and lemm.ee myself, you might be closer to right than I initially thought. The amount of comments in the first thread on the blocking instances, lemmy.world and beehaw, is much less than the federated instances, blahaj.zone and lemme.ee. The vote amounts on the blocking instances agree with each other and the amounts on the federated instances agree with each other.
I think that’s sufficient evidence to conclude something is up. I’m going to suspect lemmy.world and beehaw filtered out hexbear comments and votes when they federated the post. This would suggest brigading from hexbear users. But I would need to view the vote database to be sure.
I don’t think this is the proper way to do this analysis. I believe the lemmy.world vote counts should be the same as blahaj.zone. It will still include the votes from hexbear. The only difference would be that lemmy.world’s view will include the down votes from lemmy.world itself. Those won’t federate over to blahaj.zone since down voting is disabled on the instance.
It should be possible to do your analysis though. It will take getting a copy of the vote history for the comments on the post. Every admin of an instance that federates with blahaj.zone has a copy of that. Then you will have to run some queries on the database to filter votes by the instance they originate from.
This is the meta community. It’s actually called main song you’ll find it at !main@lemmy.blahaj.zone. Meta is being used here to mean discussions around blahaj.zone itself. The op wants discussions about blahaj.zone that happen in this community to be limited to members of the server and exclude those from others servers.
You are kind of hitting on one of the issues I see. The model and the works created by the model may b considered two separate things. The model itself may not be infringing in of itself. It’s not actually substantially similar to any of the individual training data. I don’t think anyone can point to part of it and say this is a copy of a given work. But the model may be able to create works that are infringing.
That is not actually one of the criteria for fair use in the US right now. Maybe that’ll change but it’ll take a court case or legislation to do.
NPR reported that a “top concern” is that ChatGPT could use The Times’ content to become a “competitor” by “creating text that answers questions based on the original reporting and writing of the paper’s staff.”
That’s something that can currently be done by a human and is generally considered fair use. All a language model really does is drive the cost of doing that from tens or hundreds of dollars down to pennies.
To defend its AI training models, OpenAI would likely have to claim “fair use” of all the web content the company sucked up to train tools like ChatGPT. In the potential New York Times case, that would mean proving that copying the Times’ content to craft ChatGPT responses would not compete with the Times.
A fair use defense does not have to include noncompetition. That’s just one factor in a fair use defense and the other factors may be enyon their own.
I think it’ll come down to how “the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes” and “the amount and substantiality of the portion used in relation to the copyrighted work as a whole;” are interpreted by the courts. Do we judge if a language model by the model itself or by the output itself? Can a model itself be uninfringing and it still be able to potentially produce infringing content?
Ok I can see how that could make a drama thread worse.
Only other workaround I can think of is a setting up a second instance that only federates with blahaj.zone. But that wouldn’t show up on the local feed and there may be federation quirks im not aware of.